Hallucination Evaluation on CRC-100K
1.019CHAIRsRandom
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| RandomModel=LLaVA, Shot=1-shot2026.05 | 1.019 | 0.732 | |
| RandomModel=LLaVA, Shot=3-shot2026.05 | 0.996 | 0.725 | |
| kNNModel=LLaVA, Shot=1-shot2026.05 | 0.994 | 0.7 | |
| DRModel=LLaVA, Shot=1-shot2026.05 | 0.99 | 0.706 | |
| DRModel=LLaVA, Shot=3-shot2026.05 | 0.987 | 0.678 | |
| kNNModel=LLaVA, Shot=3-shot2026.05 | 0.983 | 0.681 | |
| GAUCModel=LLaVA, Shot=1-shot2026.05 | 0.943 | 0.63 | |
| MIMICModel=LLaVA, Shot=1-shot2026.05 | 0.938 | 0.637 | |
| GAUCModel=LLaVA, Shot=3-shot2026.05 | 0.936 | 0.608 | |
| MIMICModel=LLaVA, Shot=3-shot2026.05 | 0.925 | 0.614 | |
| RandomModel=Qwen, Shot=1-shot2026.05 | 0.916 | 0.64 | |
| RandomModel=Qwen, Shot=3-shot2026.05 | 0.912 | 0.633 | |
| kNNModel=Qwen, Shot=1-shot2026.05 | 0.852 | 0.585 | |
| kNNModel=Qwen, Shot=3-shot2026.05 | 0.847 | 0.579 | |
| DRModel=Qwen, Shot=1-shot2026.05 | 0.841 | 0.577 | |
| DRModel=Qwen, Shot=3-shot2026.05 | 0.829 | 0.565 | |
| MIMICModel=Qwen, Shot=1-shot2026.05 | 0.819 | 0.56 | |
| MIMICModel=Qwen, Shot=3-shot2026.05 | 0.808 | 0.548 | |
| GAUCModel=Qwen, Shot=1-shot2026.05 | 0.803 | 0.552 | |
| GAUCModel=Qwen, Shot=3-shot2026.05 | 0.795 | 0.539 |